Format

Send to

Choose Destination
Biometrics. 2017 Mar;73(1):104-113. doi: 10.1111/biom.12547. Epub 2016 Jun 8.

Nonparametric analysis of competing risks data with event category missing at random.

Author information

1
Department of Biostatistics, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, 27599, U.S.A.

Abstract

In competing risks setup, the data for each subject consist of the event time, censoring indicator, and event category. However, sometimes the information about the event category can be missing, as, for example, in a case when the date of death is known but the cause of death is not available. In such situations, treating subjects with missing event category as censored leads to the underestimation of the hazard functions. We suggest nonparametric estimators for the cumulative cause-specific hazards and the cumulative incidence functions which use the Nadaraya-Watson estimator to obtain the contribution of an event with missing category to each of the cause-specific hazards. We derive the propertied of the proposed estimators. Optimal bandwidth is determined, which minimizes the mean integrated squared errors of the proposed estimators over time. The methodology is illustrated using data on lung infections in patients from the United States Cystic Fibrosis Foundation Patient Registry.

KEYWORDS:

Competing risks; Cystic fibrosis; Missing event category; Nadaraya-Watson estimator; Nonparametric estimation

PMID:
27276276
DOI:
10.1111/biom.12547
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Wiley
Loading ...
Support Center